| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.72, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 6.523701295013192, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0448, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.412973977164346, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0467, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.942737417685276, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0251, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.825281243026985, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9386, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.03428531076452, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9352, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.045446410621582, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9403, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 3.9268666944127144, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9175, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.8777391180341225, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.9033, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.160425946556955, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8664, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.7535355364876621, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8286, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.377354687257498, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.7985, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.3857057131138493, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7704, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.2864126093432138, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7443, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.9412049777110468, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7413, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.0651372253274827, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7158, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.0367674041378347, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7305, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.8358081831584893, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7166, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.9846595515579288, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7247, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.7850820088590341, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7018, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6904446831475367, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7021, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.8030302735307364, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.7299, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.7679704130354906, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6626, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.7462060806895223, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6272, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.8182808209893684, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.644, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.7054707552177533, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6392, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.9822291658097587, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6343, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7421621378642262, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.63, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.803064636807163, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6204, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5952083662441091, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6369, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.6504393355916186, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.607, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.6178802749932635, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6008, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.5325563902309156, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6009, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5249910781532764, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5663, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.4374253334614309, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5781, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.6136355448916558, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5751, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.41974427927769087, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5496, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.5340475975821265, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5632, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.4321640137053109, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5503, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.4456855931814425, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5486, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.43538326506767233, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.541, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.38836302885932866, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.562, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.44808851918389647, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.544, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.39486977049244426, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5224, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.38237167455272036, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5301, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.39101308084224673, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5199, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.3688767873587531, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5145, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.45187326508140646, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.528, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.37151154049333246, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.4975, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.3162156736227971, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5102, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.3328487198839056, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.4903, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.34205871153103834, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5122, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.3237384261131164, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5032, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.36046681890607807, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4815, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.3526234762023227, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.4752, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.3455590723384126, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4722, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.32051472507204953, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4932, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.32765746786469063, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4897, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.2856812281458495, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4931, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.28071386262872144, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4597, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.3149630888417075, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.4877, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.30341516554753045, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.5082, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.311393149431681, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.4839, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.2766167124366436, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.487, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.2956612605429707, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4709, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.2792838973754806, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4792, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.25002786556643203, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4663, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.2734860431392467, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4907, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.2640895932487677, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.475, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.26571488236407254, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4751, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.25059464215200034, | |
| "learning_rate": 0.0, | |
| "loss": 0.4854, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.562969851521925e+17, | |
| "train_loss": 0.6258056657654899, | |
| "train_runtime": 8871.5305, | |
| "train_samples_per_second": 0.789, | |
| "train_steps_per_second": 0.008 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 70, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.562969851521925e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |